 Hello, this is SART for an LS N-TAP. We're going to get started here in about a minute and a half. We've still got people logging on here. So we'll get started at about two minutes after. Two quick things. One of those is we've got a new LS N-TAP website, which is up. We're going through a feature improvement. We're going to be working with Urban Insight to add some features to the D-Law template. If there's anything you would like to see improved about the website, let us know. As with all of our trainings, this training is going to be recorded and it will be posted to our YouTube channel. We are just dropped into the chat there. A link to our YouTube channel, we've got over 200 legal services videos. And so please check some of those out. Let us know what you think. And if there's anything that you want to see for an upcoming webinar, we are about to do our webinar schedule for the entire rest of the year. So we are very open to any suggestions. We've got a survey and I will be dropping a link to that survey into the chat here. And we'll just pull it up really quickly right before we start. A link is coming here in just one second. I accidentally closed the window down that I had the extra link in. I'm super happy to have our presenters here today. And there was a five minute version of this talk that was given at the Legal Services Conference a few months, or last month, just barely last month. We are talking to LSE and maybe adding that mini version also up onto our website as another way to promote this stuff. But I'm turning it over at this point to Laura Quinn. Thank you so much for coming out here and doing this presentation. There's always challenges in figuring out what is really the return on investment and what type of impact projects have. And this is a very creative and interesting way to look at that. So thank you, Laura. Great. Thanks so much for having me. So yeah, absolutely. We are going to be talking through what we're calling the great equation for access to justice. So this has been a project with a bunch of partners. And in fact, you can see even more logo. It's just a second. It's been spearheaded by the Florida Justice Technology Center, of which I have been the director of data. It's been had some funding support by LSE. IELS at the University of Colorado has given a little bit of funding. And obviously, LSE and TAP is supporting us today. So what is this thing? I'm not going to figure out how to advance my slide. So this diagram is a diagram we'll spend a lot of time looking at. But in general, the idea of the equation or a kind of a framework is that or a model, I'll also call it, is that it allows us to think through quantitatively access to justice projects, even when we don't have all the data. So it's actually specifically designed to kind of hold placeholders or guest units for things that we may not have complete data for, which is an awful lot in this space. So we'll come back to this diagram and talk a lot about it. And the process of designing this has been a careful one over the course of about a year. So it's been in partnership with a bunch of research partners with on the ground we lay organizations and more. So it's based on a original Drake equation, which was created in the 1950s for an entirely different purpose. It's named after this guy, Frank Drake. And it was intended to or designed to be a kind of a thought experiment to look at, I don't know why my slide keeps advancing, to look at the number of civilizations that we can communicate with. So basically as part of a hunt for extraterrestrial intelligence. And it's just a string of numbers multiplied by each other. So the number of civilizations we can communicate with is equal to this number times this number times this number times this number, etc. There are seven factors. And some of these factors are really very knowable by science. So it starts with the number of stars that are created per year. And that's actually a number that science has a pretty good handle on right now. The next one in is the fraction of those stars that have planets. It's kind of out of my control, apparently, it just happens. The fraction of those stars that have planets. So, which is also pretty knowable. This is generally thought to be about to per, oh, sorry, that was kind of a star. I don't know what that one is on top of my head, but it's known. Others of these get very metaphysical, get very difficult to estimate. So the fraction of the planets where life develops, all we have right now is, we've got a sample size of one. We only know that in fact, well, we've got a couple of planets around the sun. So we have those to look at. This is literally going by itself. Oh, it's because of the lava fire. Okay, we'll get out of this soon. I'm sorry, I forgot to take the timing off. And so some of this gets extremely metaphysical. So what's the average number of years an alien civilization sends signals into space? Like how would we even begin to estimate this? But it's been really useful for more than 50 years. It's helped people to guide conversations. It's focused research and more. Basically, it puts a framework around if we knew all of these things, we would know the answer. So the idea of the applying this to access to justice has been to find a method by which we could take this approach to say, all right, if we knew these things, then we would know the impact of these projects. And so here it is. Here's what we've come up with. So it is also just a series of numbers multiplied together. So in this case, they kind of serve the funnel. So you've got the number of people targeted by your project at the top. People who are able to use it. I'll just let it step through because it's going to step through all of you. The number of people who found it. So basically the number of people who are able to find your site. And we're able to, in fact, do something with it, which is underneath it the received benefits. And the ones up front are generally the ideas that you are, that you could probably define before you start your project. So you've targeted it. You know how many people you've targeted. You can calculate how many people use it. As you go through, you're going to find that you're going to need to bake them in to the project. So how many found it, how many received benefits. And then at the bottom of the pyramid, whether it had a positive impact or not, or an outcome, is going to often require specific proactive research, or it might just be some guesswork right now for us to come back around to as a sector down the road. There's also this idea of financial impact. So in fact, if you thought of that, and I'm going to give you specific examples of all of these things. So once you think through the number of people who received benefit from your tool, then it's actually a fairly straightforward step to say, okay, and what was the financial impact of each of those people serve so we can actually come to the potential dollar impact either savings to the sector. So they're savings to legal aid, savings to the actual person who needs the help, other things like that, savings from health care. All right. So and all of these things are we so we use numbers where we have them. We use hard data where hard data exists. But in cases where we don't have hard data, we use proxies. So we use places where we say, all right, well, this is exactly the same, but I'm going to say it's similar enough to use. So for instance, to say, all right, well, we don't know exactly how many people are faced with conditions issues and issues in their rental housing, but we suspect that it's going to be similar in number to, let's say, the number of people who are living or have a rent burden of 200% something like that. So basically we say, all right, this isn't exactly the same, but we we're going to base it on something or we can just make a plausible guess. So we can say, all right, well, I don't actually know this number, but I'm going to guess and it basically then it becomes a thought exercise to say, all right, what's plausible? What's the plausible impact of it? Let's look at we're going to look at a couple of examples. I'm going to walk you through. So at the same time, I'm going to walk you through this example. I'm going to walk you through a fictitious example of a kind of modeled on a statewide website, and I'm going to walk you through the worksheet that that we've developed that will help and I'm going to move away from these slides because that's definitely not where I'm going to walk you through a worksheet that we've developed to help you think it through yourself. So any questions about kind of the whole idea at a high level before I start diving into kind of how this idea is applied? So any questions into the chat? I love questions. Otherwise, I'm standing here in my office by myself talking to myself. Start. I'm not trying to be able to easily see the question. So just to remind people, there are two ways to submit questions. One of those is to use the question box that is on the control panel. I monitoring that and can read those out. The other one is to use the raise your hand function. If you do that, we can unmute you and you can ask questions that way. But we're given kind of the size of the audience. We're willing to have a conversation over the this if anybody has questions. We don't have any at this point yet, though. Absolutely. Great. So our first example, we're going to be walking through actual an actual model or the Florida name change website. So this is an actual website created by Florida to help folks update their their name and their gender marker. So it's specifically targeted at the transgender community for whom it's often a big barrier that they've transitioned their physical appearance. So, for instance, they've gone from male to female in physical appearance, but they still have an ID that has a name, which is in congruence and gender marker. So it says they're female instead of male. I think it was my first with my example, male set of female. And that can just lead to lots of discrimination and just difficult times harassment. So let's let's start by looking at the the top of the funnel. So looking at the the first couple of steps for Florida name change. So looking at the number of people targeted and for how many of those folks is the site accessible. So for Florida name change, we started with the number of adults who identify as transgender in Florida. And we have we actually have a couple of really great reports, which made this a particularly great example. So this is actually a I'm going to show you another example. Where the data is really is good, which is a probably more typical example. So but here is an actual number from Williams Institute as from adults who identify as transgender in Florida. So we then need to think about how many of those folks who are transgender have actually have made a appearance transition from one gender to another. So not all of them have. Twenty five percent is what is estimated or what's been reported by this other report, which is actually a fantastic source of information for the whole thing. And then we need to think about how many of the folks who have transmitted transgender parents actually wants to change their driver flight into the state ID. And this is the number of people who expressed that desire. One of the things that we need to think about as we look at this stuff is whether how related these two things are, like, for instance, we've just put both of these things and we've said, OK, of the 25 percent who identify as transgender, only 58 percent of them want to actually change their driver's license. If these actually are going to overlap a lot, then we might need to think more about that. So if, for instance, half the people is almost all of the people who want to change their driver's license, how, in fact, transition their gender, then that's going to affect our bottom line. Let me show you a different example. So this is Springfield Legal Help. This is a fictitious example. It's based on the model of, well, I guess it's a VA, not a statewide website, but a city website, legal website. So looking to provide kind of core legal info for low and low and middle income people. So here we're looking at the number of people targeted and we might say, OK, well, we want to look at the number of people in Springfield who are below 300 percent of poverty. So here's our core number. And then we're going to factor that down to say, well, not everybody who lives in poverty has necessarily a civil legal question. So let's say 40 percent. This is based on a model we're still working on from Florida, and we're still getting down our swords for that. And then this actually came up for us. This might not be the same for every website like this, but it became important to us to try to define to say, OK, we're not trying to answer all of those questions with pro se materials. We feel that some of these civil legal questions are really only answerable by a lawyer and we're not going to try to take them on. So that was 60 percent that by a careful estimate is kind of was the target, you will say, and it's a fictitious example for this site. So that leaves us a number in our target audience. Let me just show you how this plays out in this magical worksheet that we've created for you. I'm going to show you this first and then we can show how it moves down. So this is a worksheet that's designed to calculate through and has a ton of instructional text. The reason that it stands with draft all over it is that it is really we need to get it out in the field and people working with it in order to see how well it works. Like, for instance, I'm worried that this may be a overwhelming amount of instructional text and examples. There's more example here. But possibly more is better than less with something that that's somewhat complicated. So so the way this works is basically you would start with the number of people or households in your geographic area. It's probably going to be your typical starting point. It's likely going to be from census data. And then you're going to think through your limiting factors and your limiting factors are what we just looked at for these other examples. So this would be, you know, what percentage of them have changed their physical appearance if you're looking at transgender? What percentage of them actually have a civil legal question? What percentage of them are living in poverty? And you might have more than one limiting factor. So you're writing them in and you're putting your data on your data source when you write it in. So we're going to say there is thousand people in our target audience and this one's 45 and this one's 67 and this one's 89. It's going to automatically calculate for you. And then this calculation automatically populates this summary tab, which goes with the diagram. So basically we've got number targeted, which is all we talked about so far is the top level in our diagram. And here we've got number targeted and this is calculated out for your project. Give me questions in the chat. In the meantime, I am going to move on and I'm going to talk about some examples for percent accessible. So the next level is thinking about how many of the people we targeted can actually get access to this information. And it's a little wiggly as to which of these are in which so whether the, you know, like, for instance, are we targeting only English speakers and so we're going to take them out here or are we going to say, well, really our ambition is to support everybody in the city of Springfield or all transgender people in Florida. And thus we're going to take them out here. So we need to acknowledge somewhere that people who don't speak English can't use our tool or people who can't read English can't use our tool because it's, you know, written information and it's in English. But it's kind of a matter of so you could say that's not our intent for the tool. We don't intend to support Spanish speakers. In this particular case for Florida name change, that would be a little uncomfortable. It would imply that we only care about English speaking transgender folks and not Spanish speaking transgender folks, which I don't know why that would be our strategy. So so in this case, we have not said that. So basically this is the percent of the population that can to actually read. Sorry, I messed with this a little bit. A brand and this is not our live model. So this is the percent of the population that can read the language that's are included. It's important to for any written thing to include literacy. We've been kind of blown away as we are looking at this, how in fact I have it in the next one, how low literacy rates are for a lot of ways that you might measure them. And how many people are able to use written information on the Internet? Oh, actually, here's what we did here. So we have how many people can speak the language? And this was actually up at 90 percent, something like that. And we basically said how many people are able to use written information, including through help, through sources of help. So for instance, somebody who is transgender and wants to change their gender marker, how many of them can find somebody to help them use this tool? And this is a guess. I strikes me it might be kind of high by that definition. But basically, we're we're taking a combination of, as you can see, data from like the census survey and kind of some some things that are really hard to know. We actually did a deep dive for a different estimation project on trying to actually estimate how many people had access to the Internet. And it is surprisingly difficult. So we kind of punted with had access to the Internet or to find help to do so. And that means the that together, 88 percent of the folks that meet in our target audience can actually get access coming out to 12000. Springfield Legal Health. So what percentage can read English at a basic level or above? This is definitely take a look at this for yourself. Things get really kind of alarming when you're when you're looking at people, what you define as basic, what you define as being able to read. So basic is about a I think we looked at it. There's not a good definition. But this was about a fourth or fifth grade level we felt. What percentage of the audience can get info online or can get help to do so? So this is our 95 percent number. So then we can't just multiply these numbers by each other because we have to assume that a lot of the people who can't get online are a lot of the same people who can't read English at a basic level. So they're they're not going to be completely multiplicative. So basically you've got the very technical term. Here's an eyeballed mashup of the above stats. So basically this is saying, all right, it's close but not quite multiplied together. And then here's another really important one to think through for content sites. So of so we have, you know, there's a lot of people in Springfield that got lots of civil legal questions of all of those questions. What percentage does our site actually answer? And for almost all of our content website, the answer is probably, you know, not a vast majority. You know, like there's always a lot more questions that one can ask. So it's useful to estimate, OK, so by volume of number of people asking, what percentage have we done? And this 30 percent would imply that in Springfield we're kind of just starting out here. So we haven't gotten very far through the things that we think we'd like to get to eventually. And this means that if you take these numbers, so the number of people who can read it and the number of people who can use it to answer those specific questions, this results in a remarkably low number. And this is kind of a useful, you know, this is one of the places where we can start to see the value of doing a mental exercise like that, even if we're guessing some of these numbers. Because, wow, if only 30 percent or only 90 percent of our target audience can use it, should we be thinking of doing other things? Like what could we do? Like maybe a Spanish level, a Spanish version of it, maybe adding videos to the site to target people who aren't literate for written things, maybe adding more legal content. So we can start to think of that. So here's our total number of people who can use it. I'm just applying this to our worksheet. Sorry, I lost it for a second. Here it is. I'm a number able to use it at a moment of fear when I thought it wasn't in here. So this tends to be relatively straightforward. We found a cross-project. So we've just dumped the key things that we have found to be useful. So what percentage of the population can speak the language that are included? What percentage of the population can read and access the internet? What percentage of your target audience is likely to want to know? What is likely to want or know to do about your topic? So basically, what is the content that is available to address your target audience needs? And if you fill out those numbers, they will populate. Let's do some because it'll populate not just, let's say, 90 percent. So it will. OK, so I need to prove this. So it will populate something which is not the point three people who are able to use it. But it'll give you both the number and the percentage. These might be backwards. Yes, there's a backwards. So it's 94 people in our target day. I'm starting with all the people. 94 people to whom the site is accessible and 35 percent. That's 35 percent target audience. Onward, let's people have questions. Number of people who found it might be actually the most straightforward level here. So this is simply like it's found. How many people actually went there at all? And given if we're saying that this is specifically online tools, online tools, it's generally fairly easy to measure who came through something like Google Analytics. We've got something unique visitors is generally a good number. We need to figure out how many of those folks are actually in our target audience. And depending on who your target audience is, that may be straightforward. So if we were targeting everybody in Springfield, that would be easier because we're targeting people who are below 300 percent poverty. We need to actually ballpark how many of these folks are actually 300 percent full of poverty. So and in this case, we usually benchmark a couple of other sites to see what would make sense. So like, for instance, in Illinois, 90 percent of the people who use the Illinois Guide and Help are below 300 percent poverty. 80 percent are below 100 percent of poverty who use the Massachusetts Guide and Help. So that given those statistics, it felt like 65 percent was a pretty conservative estimate. And basically when you're doing ballparking like this, we've very much kind of defaulted to, well, we're not really very sure. So therefore, let's default to something more conservative. And then it's more defendable at the end when you say, all right, it's this many people impacted. It's this amount of financial impact. You can kind of say, well, given these extremely reasonable guesses, this is kind of the minimum impact. Here's the percent that are actually from, so let's say it's Springfield County. Google Analytics is kind of a pain to get out city. Data, in fact, you really can't. You'd have to parade it, which I could say more, but I won't. So basically just know that you're not nuts in thinking you can't get out city level data from Google Analytics or zip code level data. So and then thinking about a potentially the number of page views who found this particular site. So this has just people. The page views is not actually in this calculation. So this is the number of people who found it. And then we can go ahead and look at things like the percent of those accessible who found it. So which is kind of an interesting statistic. So of all of those folks who could use the site, how many used it? In fact, what we find is that a lot, you know, a third of all people who could use this site have used this site. And thinking about it in this way actually gets to be fairly interesting because taking out, like, for instance, the accessibility issue is really helpful because you could also if you just look at this and say, well, what percentage of all of the people who targeted have actually come? Well, that's only five percent. And that looks a lot. Well, I mean, it looks a lot worse. But in fact, well, I don't know if it's worse or better. In fact, we've done it to ourselves by not making it actually usable by the people not usable for at least what they're hoping to use it for by the people that we're looking to serve. Let's look at Florida name change. A number of people found it. So it's a very similar concept. So the number of unique visitors from Florida, we just went ahead and go on those into one line. We need to then try to limit it to transgender folks, which we don't have. We don't actually ask folks whether they're transgender. So there's no easy way to know that. So what we did is we looked at the percentage of those assemble at least one form that includes a change in gender marker. So this is actually going to be a very low statistic because you could imagine that there are some folks who are kind of just going through the first step and they're changing their name, but not their gender marker yet, who are actually transgender. But so we have a low estimate of the number of folks on our site who are transgender. And this is so comes out to about 23 percent, 24 percent of our target audience. This is actually a slightly different statistic. This is actually of the target audience, as opposed to those of who is accessible, I think. Yes, who have actually visited the site. Both are pretty useful statistics. And I think I put them both on the worksheet. So basically, in fact, a quarter of all people who we are targeting have visited the site, which is pretty impressive. And there's been a crap load of promotion around this tool in a pretty tight and connected community in Florida. So it's I think that this is a reasonable number that we've actually hit that many. All right. And in the worksheet. So we're going through tabs here. So there's one card, a tab for targeted and then also be able to use benefit, sorry, able to use level. Then another one for found it and the actual benefits. So I'm going to go through this and then just to have a little bit of a change of pace. I'm going to I'm going to put Brandon on the slide and I'm going to ask him just kind of anything that he's found to be particularly interesting as he's kind of gone through this exercise, thinking about it specifically from Florida in this founded level, kind of anything to add. So I'm going to go through the spreadsheet first, Brandon, so you have a minute, take your time in order. So this is probably a fairly straight forward. Well, not necessarily straightforward to know, but at least straightforward to think about what the numbers are. So this is an easy one, probably the number of unique visitors is likely from Google Analytics, the number of people specific to your geographic area. So thinking through how you move from everyone to your geographic area and then thinking through what percentage of this target graphic is from your target audience. This could be a guess. So if we weren't doing document assembly, but we're targeting transgender folks, it would be pretty darn difficult to know whether folks were transgender or not. So we would try to have to we would have to do some kind of guess as to how many those were. All right, so if we say we've got, actually remember my numbers keep them separate. So let's say we've got 300 people and 250 are specific to our geographic area and 90 percent are from our target audience, 225 of them found it. Then this populates my diagram. So here's my number who found it. And in fact, this is embarrassing. I don't go through and fix the worksheet before we disseminate the worksheet. I thought I had tested it really well, but I had not. So and I might actually do two two numbers here. So the percent of targeted who found it and the percent acceptable found it. They think both of them are useful. Brandon, so having been through this exercise for a couple of slides, any tips on kind of where we've the levels we've been through so far? Honestly, well, first of all, I'll say it definitely is it's a very amazing framework if you guys kind of get a chance to work through it. Laura has done an amazing job of actually putting this all together. A lot of thought to it. I don't know. It's kind of interesting thinking about the funnel. There's just there's just a lot of things that I think are more eye-opening about it than anything. A lot of the things we've been struggling with when we've been doing our models as far as the readability statistics, you know, how many people can read the basic levels and things like that. And I think by going through the process of thinking through this, you really do see where some of the pitfalls are and you understand that there are some things that may not, you know, work as well for your users as you think they are. So I think that that's one of the things that's been very eye-opening to me. And then the other thing is we've kind of been doing like a lot of the impact, trying to figure out the dollar impact of a lot of our tools. You know, we found that there probably are, you know, ways that we can work, you know, better as a community to kind of think through, you know, how do we value like a lawyer's time or something like that on these types of cases or how do we value, you know, the type, type or the amount of legal aid that we're providing to people. So I think there's a lot of interesting questions, you know, on the impact side that, you know, as a community and I kind of just to give you, I guess a little quick plug, Laura, we'll talk about it later too, but we're thinking about having an ongoing group, you know, for the Drake equation that as a community, we can think through a lot of these things and kind of help each other come up with better, you know, better reasonable estimates for these type of, you know, figures as we're kind of working through them. And I mean, that was kind of my overall thoughts. I mean, if you work with it, it is really eye-opening. There's a lot of things you'll see that you probably weren't expecting and it can really help you understand your site, you know, and your visitors and how you can, you know, serve them better. Fantastic. Thanks, Brandon. In fact, I can introduce Brandon. The new director of data and actually several other things at Florida Justice Technology Center who's been working with me and applying this to Florida stuff. So great. So let us continue to delve down the funnel here. So we've got then the number of folks who took action. Actually, this is change names. This is the number who received the benefits, received the benefits. So this is where we start to get a little more interesting than simply the number of people who came to the site. But we don't yet have to worry too much about, all right, has it changed their life or not, which is down in the positive outcomes. So I think it's useful, very useful to separate the two, because I think we can often get ourselves really wrapped around the difference between, for instance, outputs and outcomes for those of you who are familiar with that model or whether something is really a significant enough benefit to count. So basically this model says you could essentially you can count anything you want. And the worksheet allows you infinite columns to count whatever benefits you think are useful. So here are the things that we've... So in fact, for the content website, we kind of worked through a model with just gets to one bank and you could have multiple. But this has just one, which is you're trying to get through how many people understood their options after reading this page. And so that's obviously not a incredibly easy thing to know. We stepped through a couple of different things to try to get something that we felt was plausible. So well, something that we know is that not all visits to the site are visible content page. So logically, they're not understanding their options if all they've used is a home page. So we can read those out. We have scroll data on the site. So we can say, all right, who did any scrolling at all on this page? And this is a proxy figure. So it's possible that someone completely understood their options without scrolling down the page. But it's certainly just as likely or more likely that people scroll down the page without understanding any options. So this is a we basically use this as a proxy to actually do we basically pay someone who engaged with the page who actually looked for information. And then then there's people how many of those folks who actually look for information actually read enough for read enough of the page to actually get their answer. This is a guess. We are there is data coming out of Ohio fairly soon. We actually just done a a randomized controlled trial to try to get some numbers around, you know, to what extent it does do. In fact, does getting people information actually help them to understand options and encourage them to take action so they'll be more sooner. But then coming down to the percent that understood their options after reading the page. And so if you multiply all of these together, this comes to 13 percent of those who found the site were able to understand their options for their question. And a lot of this is simply guesswork. The one of the ideas of the Drake equation is well, so a couple of things. Number one is that a guess is better than just kind of saying throwing up our hands and having no nothing at all to estimate with. Number two is with is that if we get together as a community to have similar guesses across the community, at least we'd be baselining based on the same thing. And then that would also number three allow us to do research to allow us to improve those guesses over time, which is something that's the original Drake equation for extraterrestrial life has really proven to be good for. So looking at these numbers, although we might look at them and say, well, this is I mean, this is nothing but guessing. These are this is stuff that is researchable and we could get to eventually. So here's what this looks like. So this is. I think this received benefit. So this is because of the document assembly site, this actually becomes very straightforward for change. So basically the we because there are a bunch of people filling out multiple documents, we we just pulled a single one that is the most filled out with a gender marker as the proxy for how many people are doing things. So this is a low number for how many people are doing things that some people might not do for screen card. And then the number of actions we were interested in just looking at the funnel here of how many people actually completed it. So we looked at that. So here is the number of people who completed. Oh, sorry. This is the number of people viewed it. And here's the number of people who completed it. So then we've just pulled down the number of people who completed it as our our benefit. And we can then again, we can look at this as a percent of target visitors. So seven point one percent of people who came to the site received the benefit as we find the benefit. And here it is in the worksheet. So this is essentially there's not much to put in in this spreadsheet because the benefits are really going to vary by project. So so you're going to have to think through what measurable things you want to find as benefits in order to put in this spreadsheet. So if we actually so these these just some so 32 so 10 people received this benefit 15 people received this benefit 57 people received benefit. This is plugged in here. All right. Alcoms. Well, I'll start on this one because it's most tactical to be able to see it. But it's almost unfair because there is a fantastic report that allowed us to have data that we could actually use there. There is a report that actually estimates or it has how many people how many transgender folks have recorded being harassed or worse based on having an ID that doesn't match their their physical gender appearance. And so we have statistics for this. So 32 percent of transgender folks have been been harassed or denied service or attack. Nine percent, so more specifically denied service are at the lead. Two percent have been attacked. And so we can calculate this through and say, all right, here is the number of people who were just spared harassment. So this is a total number. And then we have more specific number, including, you know, a pretty powerful four people who are spared attack based on this. I'm going to go all the way through to the the financial impact on on this because it's pretty related. You could look at a number of different things to measure a financial impact. Like you could, for instance, estimate the amount of time that it would take someone to do this pro say and be and put that at minimum wage or something like that. In our case, lawyer hours is for pro bono attorneys or sorry for legal aid attorneys is kind of an easy thing to to use. So we have we've estimated that it takes 1.75 hours to do this whole process to fill out all the forms needed. And so if we say that it's also if we take the one rate of a legal aid attorney, not what they actually would bill, but what they're actually costing the legal aid sector is $90 an hour. Here's the savings and money that's been produced. All right. So positive outcome for our Springfield legal health just gets super wiggly. So very let's put a finger in the air and get something. So how many so we've said, all right, you're the number of people who actually understand their option. How many people have done something or correctly not done something that has made a positive change in their life compared to having no information available. So this is probably one could go down a rabbit hole into like behavioral science and find out how many people are likely to take action based on having information. I have not done that. We haven't done that yet. And so this is just a, you know, kind of out of out of nowhere guess. But it's presumably, you know, it's probably going to be at least a few percentage points. It's definitely not going to be 90 percent. And so we hopefully as a community can get closer and closer to that. So here is then a plausible number for whom Springfield legal health has has helped move in an actual productive life direction, which is not an impressively high figure right now. This is not necessarily that we get Springfield legal health, which doesn't exist, would want to share with a funder in exactly this form. But one of the things that you could do with it is you could say, all right, well, how much, for instance, is like we can use it as a model to say, all right, well, if we said that 90 percent of our target audience could actually use this information, how much does that change it? Oh, it doesn't because they're not coming. So we'd have to figure out, we then have to model how many more people do we think we actually want to change this to say, all right, how many how many people are finding it because it's now accessible to them? So yes, you can use it in that way. And here we have so we could say, for instance, if they understood information, it was equivalent to talking to an attorney for, let's say half an hour because attorneys don't talk for less than half an hour. So and for no reason, we use a different number here that we can say, all right, so our one rate is $90 and that equates to about $50,000 of value. So and in fact, this is a much more impressive number than this. And neither are going to probably support the program unless, you know, you're making a big case for why it's doing really important stuff in other ways or how you're improving it year by year. But basically it's because this assumes that many of the folks who talk to a lawyer also did not make a positive change in their life, which is probably true. So basically this is the financial value of having saved the lawyer time. Great. I'm playing that out in the worksheets. This actually, this worksheet looks a little different to try to allow you to have a bunch of other I have a bunch of outcomes if you want. So basically the idea here is that I'm going to take the benefits. So things stuff from here. People who received benefits, I'm going to write in the benefit. And I'm also going to write in the benefit here. So sorry, my draft is on top of everything. So my benefit might be created a danger form. And then I'm going to just fill in this is already filled in. It's pulled over from the other sheet for me. The people who did benefit number two. So we then have we then estimate OK. So we say what is the outcome that we think is happening. So basically for this example that I'm walking through here. So I'm doing game change. This might be saved harassment. And we say that whatever that number was 29 percent. We saved from harassment. And then this is how many people that actually was. And then you could do something else. Saved. Poor people and that is a fraction of a person. And you can do the same. So this is the model here primarily works with the assumption that you're saving somebody hours, which is not the only thing that you could financially estimate, but is a good guess. It's probably a place for a lot of people to start. So you can go through there. And this worksheet throughout has had example cases that follow throughout this last tab simply has these example cases filled out for your reference. A few last closing thoughts. And I would love to take questions, thoughts. Anything you have to close. So we would love any or all of you to apply this to whatever makes sense to you. This is all three to use to do whatever you want with to modify, do whatever. So we have the recording. So we'll have a recording of this. We have a recording of the the five minute speed talk I gave at TIG. The worksheet, which I'll now update, so it's actually correct. We'll all be at this URL online, which I think SART will now also put into the chat. Think about not only the obvious application for evaluating your projects, but think of it as think of it potentially for comparing projects to each other to create baselines to think through it like before you start a project to look at a bunch of other people who have done projects like it to understand what this looks like for them, which would help you understand what to expect and kind of what baselines you can use. And as Brandon mentioned, we're eager for anybody who is who is interested in applying this to their own projects to starting a group of people who will simply discuss what they're doing and work together to do things like developing shared estimates on important things. So actually, you can let us know in the chat right now. If you know you're interested in joining a potential group, maybe like a monthly phone call about this or you can email Brandon at Florida Technology Center dot org. So one questions come up. What are the common estimates or guesses that you think really need a good guess so that we can compare things across projects? Thank you. I would say that the ones that I would most like to really understand more of are. So when we think about so there's so many projects are about reading information or about taking action after somebody has done things that thinking through how many people who go to a page actually understand the page, how many people who understand the page actually take action? How many people who create a form, document assembled actually file the form? So things that are, I think, pretty tangible to be able to say, all right, people are taking just kind of one step beyond what we can easily measure. How do we guess or do research to to get to that? Awesome. Other questions, other thoughts? If you were going to apply something like this to an online intake model in some way to kind of see how many people are actually finding the services that might be eligible and then what are some of the other things that you would really want to know in order to look at this? Because online intake is one of the most popular areas for grants and tries to speed up the access for individuals. Yeah, yeah, absolutely. So I think the number targeted and the able to use it piece are straightforward. So number targeted is the number of people that you so actually it gets interesting. It's how many people it's possibly how many people would you think would be eligible for your services or it's some it's that plus people who might wonder whether they're eligible for your services. Like it's probably both of those together, I might suspect, especially if it has a triage element. So you want to think through that. So who are you actually targeting? You're targeting only people who are eligible or are you targeting kind of a larger population to help people understand that? That able to use it is going to be very similar to other things that we've talked about. You'd want to think through who can actually get online, you can read it, you can speak the language. Found it would be a straightforward one. That's how many actually hit the top of the intake so that they've come to the intake. Received benefits is where things get really interesting here. So you might have a couple that are really important. So logically, number of intakes is going to be an important benefit. So the number of people who actually were feared for a phone intake. But you may also have people, like the people who made it all the way through some kind of triage, but were referred to somebody else or some other information. That could also be an important benefit. Those are people who fundamentally are not calling the line and thus are saving hotline staff to do something else. So that might be another important benefit. And then positive outcomes. So if you're thinking of people's lives, then you'd ideally want to be looking at all right and what kind of service did those folks receive? Were they more or less likely to receive service than the folks who did not go through the online intake? Were they more or less likely to get a whatever your organization defines as a kind of a good outcome if you're right hearing that? We actually just, I also just finished up a project with Illinois with the layout. And there's a report, I think it's released or will be released soon that looked at the whole and layout. So specifically the layout online intake and triage process which actually did look at benefits and outcomes down at the end. Fantastic. Other questions? That's all that we've got at this point. Thank you so much for putting this on. We greatly appreciate it. The chat channel does have a link to our YouTube channel. This should be up within the next few days. We will also have a link to all of the resources that are available over at Florida. I am redistributing that link here. It's at the floridajusticetechnologycenter.org slash break but that URL is also there available in the chat to the entire audience. There is a lot more detail and in depth stuff there so I definitely recommend checking it out. We'll have a follow-up survey and also the webinar survey is in there. So if there's topics including this that you want to see more on, please put that into the webinar survey. We're choosing our webinars for the rest of the year here in the next few days. Thanks. Thank you so much guys, see you soon.